Safety profile of Alzheimer's disease patient populations over 18 months using ADNI and controlled clinical trial data

Abstract

Studies of potential disease-modifying therapies for Alzheimer's disease (AD) are of ≥18 months duration and involve an elderly AD population at risk for serious concomitant illnesses. These conditions can lead to death or serious adverse events, the rates of which can affect the design, monitoring, and interpretation of AD studies. Safety results were summarized for the placebo treatment groups in 5 published 18-month clinical studies in patients with mild or mild-to-moderate AD (N = 1512) and compared with safety data in the Alzheimer's Disease Neuroimaging Initiative (ADNI) database in patients with mild-to-moderate AD (N = 190) participating for up to 18 months. Among the 5 published treatment studies, the mean age of placebo-treated patients ranged from 68 to 77 years and the mean baseline Mini-Mental State Examination (MMSE) scores (3 studies available) ranged from 20.7 to 23.3. Mean age in the ADNI cohort was 75 years (range of individuals 55 to 91 years) and the mean baseline MMSE score was 23.3 (range of individuals 18 to 26). Among the 5 published studies, the percentage of placebo-treated patients who died during the 18 months following enrollment ranged from 0% to 2.4%, the percentage (3 studies available) who experienced a serious adverse event (SAE) ranged from 20% to 21%, and the percentage (3 studies available) who discontinued due to an adverse event (AE) ranged from 2.7% to 11.6%. In the ADNI cohort, the percentage of patients who died within 18 months of enrollment was 2.1%, the percentage who experienced an SAE was 21% and the percentage who discontinued due to an AE was 3.2%. The incidence of specific AEs from the published treatment studies and the ADNI cohort will be presented and discussed. The rates of death, SAEs and discontinuations due to AEs can be considerable in placebo or untreated AD patients during an 18-month assessment period. These data will be useful when determining sample size to adequately power new AD trials and, along with the incidence of specific AEs, will be useful when monitoring the safety of ongoing blinded trials, and when interpreting the benefit:risk ratio of study results.